LiDAR point cloud is essential for autonomous vehicles, but motion distortions from dynamic objects degrade the data quality. While previous work has considered distortions caused by ego motion, distortions caused by other moving objects remain largely overlooked, leading to errors in object shape and position. This distortion is particularly pronounced in high-speed environments such as highways and in multi-LiDAR configurations, a common setup for heavy vehicles. To address this challenge, we introduce HiMo, a pipeline that repurposes scene flow estimation for non-ego motion compensation, correcting the representation of dynamic objects in point clouds. We further propose SeFlow++, a real-time scene flow estimator that achieves state-of-the-art performance on both scene flow and motion compensation. We validate HiMo through extensive experiments on Argoverse 2, ZOD and a newly collected real-world dataset featuring highway driving and multi-LiDAR-equipped heavy vehicles.

The capability to efficiently search for objects in complex environments is fundamental for many real-world robot applications. Recent advances in open-vocabulary vision models have resulted in semantically-informed object navigation methods that allow a robot to search for an arbitrary object without prior training. However, these zero-shot methods have so far treated the environment as unknown for each consecutive query. In this paper we introduce a new benchmark for zero-shot multi-object navigation, allowing the robot to leverage information gathered from previous searches to more efficiently find new objects. To address this problem we build a reusable open-vocabulary feature map tailored for real-time object search. We further propose a probabilistic-semantic map update that mitigates common sources of errors in semantic feature extraction and leverage this semantic uncertainty for informed multi-object exploration. We evaluate our method on a set of object navigation tasks in both simulation as well as with a real robot, running in real-time on a Jetson Orin AGX. We demonstrate that it outperforms existing state-of-the-art approaches both on single and multi-object navigation tasks. Additional videos, code and the multi-object navigation benchmark will be available on https://finnbsch.github.io/OneMap.

Preference-based reinforcement learning (PbRL) is a suitable approach for style adaptation of pre-trained robotic behavior: adapting the robot's policy to follow human user preferences while still being able to perform the original task. However, collecting preferences for the adaptation process in robotics is often challenging and time-consuming. In this work we explore the adaptation of pre-trained robots in the low-preference-data regime. We show that, in this regime, recent adaptation approaches suffer from catastrophic reward forgetting (CRF), where the updated reward model overfits to the new preferences, leading the agent to become unable to perform the original task. To mitigate CRF, we propose to enhance the original reward model with a small number of parameters (low-rank matrices) responsible for modeling the preference adaptation. Our evaluation shows that our method can efficiently and effectively adjust robotic behavior to human preferences across simulation benchmark tasks and multiple real-world robotic tasks. We provide videos of our results and source code at https://sites.google.com/view/preflora/.

Previous dominant methods for scene flow estimation focus mainly on input from two consecutive frames, neglecting valuable information in the temporal domain. While recent trends shift towards multi-frame reasoning, they suffer from rapidly escalating computational costs as the number of frames grows. To leverage temporal information more efficiently, we propose DeltaFlow ($\Delta$Flow), a lightweight 3D framework that captures motion cues via a $\Delta$ scheme, extracting temporal features with minimal computational cost, regardless of the number of frames. Additionally, scene flow estimation faces challenges such as imbalanced object class distributions and motion inconsistency. To tackle these issues, we introduce a Category-Balanced Loss to enhance learning across underrepresented classes and an Instance Consistency Loss to enforce coherent object motion, improving flow accuracy. Extensive evaluations on the Argoverse 2, Waymo and nuScenes datasets show that $\Delta$Flow achieves state-of-the-art performance with up to 22\% lower error and $2\times$ faster inference compared to the next-best multi-frame supervised method, while also demonstrating a strong cross-domain generalization ability. The code is open-sourced at https://github.com/Kin-Zhang/DeltaFlow along with trained model weights.

By combining rotary- and fixed-wing flight, hybrid uncrewed aerial vehicles (H-UAVs) can uniquely address missions combining long-range aerial transport and precise ground-relative tasks, such as the placement or retrieval of payloads. However, to leverage their full maneuverability, first, the fundamentally different operating modes of rotary- and fixed-wing vehicles need to be unified and second, the system be controlled precisely despite complex aerodynamic effects. This work presents a general and lightweight, cascaded control formulation for such versatile and accurate operation of H-UAVs. First, a novel guidance law unifies ground- and air-relative position control modes typical for the individual flight regimes. Second, we formulate a jerk-level feedback-linearization to accurately track the guidance outputs despite model errors and disturbances. In extensive real flight tests with a tiltwing H-UAV, we demonstrate the versatile allocation of (hybrid) flight states and the overall accuracy enabled by the control system. Position errors remain below 0.5 m (one quarter of the wingspan) in the full flight envelope, including accelerated maneuvers up to 10 ms2and gusting wind reaching 12 m/s. Finally, the control system demonstrates exploiting hybrid flight for transport-related missions with a precise, in-flight pickup of a payload.

LiDAR point cloud is essential for autonomous vehicles, but motion distortions from dynamic objects degrade the data quality. While previous work has considered distortions caused by ego motion, distortions caused by other moving objects remain largely overlooked, leading to errors in object shape and position. This distortion is particularly pronounced in high-speed environments such as highways and in multi-LiDAR configurations, a common setup for heavy vehicles. To address this challenge, we introduce HiMo, a pipeline that repurposes scene flow estimation for non-ego motion compensation, correcting the representation of dynamic objects in point clouds. During the development of HiMo, we observed that existing self-supervised scene flow estimators often produce degenerate or inconsistent estimates under high-speed distortion. We further propose SeFlow++, a real-time scene flow estimator that achieves state-of-the-art performance on both scene flow and motion compensation. Since well-established motion distortion metrics are absent in the literature, we introduce two evaluation metrics: compensation accuracy at a point level and shape similarity of objects. We validate HiMo through extensive experiments on Argoverse 2, ZOD and a newly collected real-world dataset featuring highway driving and multi-LiDAR-equipped heavy vehicles. Our findings show that HiMo improves the geometric consistency and visual fidelity of dynamic objects in LiDAR point clouds, benefiting downstream tasks such as semantic segmentation and 3D detection. See https://kin-zhang.github.io/HiMo for more details.

We present COIN-LIO, a LiDAR Inertial Odometry pipeline that tightly couples information from LiDAR intensity with geometry-based point cloud registration. The focus of our work is to improve the robustness of LiDAR-inertial odometry in geometrically degenerate scenarios, like tunnels or flat fields. We project LiDAR intensity returns into an image, and present a novel image processing pipeline that produces filtered images with improved brightness consistency within the image as well as across different scenes. We effectively leverage intensity as an additional modality, using our new feature selection scheme that detects uninformative directions in the point cloud registration and explicitly selects patches with complementary image information. Photometric error minimization in the image patches is then fused with inertial measurements and point-to-plane registration in an iterated Extended Kalman Filter. The proposed approach improves accuracy and robustness on a public dataset. We additionally publish a new dataset, that captures five real-world environments in challenging, geometrically degenerate scenes. By using the additional photometric information, our approach shows drastically improved robustness against geometric degeneracy in environments where all compared baseline approaches fail.

This paper presents a novel method for using Riemannian Motion Policies on volumetric maps, shown in the example of obstacle avoidance for Micro Aerial Vehicles (MAVs), Today, most robotic obstacle avoidance algorithms rely on sampling or optimization-based planners with volumetric maps. However, they are computationally expensive and often have inflexible monolithic architectures. Riemannian Motion Policies are a modular, parallelizable, and efficient navigation alternative but are challenging to use with the widely used voxel-based environment representations. We propose using GPU raycasting and tens of thousands of concurrent policies to provide direct obstacle avoidance using Riemannian Motion Policies in voxelized maps without needing map smoothing or pre-processing. Additionally, we present how the same method can directly plan on LiDAR scans without any intermediate map. We show how this reactive approach compares favorably to traditional planning methods and can evaluate up to 200 million rays per second. We demonstrate the planner successfully on a real MAV for static and dynamic obstacles. The presented planner is made available as an open-source package11https://github.com/ethz-asl/reactive_avoidance.

Micro aerial vehicles (MAVs) hold the potential for performing autonomous and contactless land surveys for the detection of landmines and explosive remnants of war (ERW). Metal detectors are the standard detection tool but must be operated close to and parallel to the terrain. A successful combination of MAVs with metal detectors has not been presented yet, as it requires advanced flight capabilities. To this end, we present an autonomous system to survey challenging undulated terrain using a metal detector mounted on a 5 degrees of freedom (DOF) MAV. Based on an online estimate of the terrain, our receding-horizon planner efficiently covers the area, aligning the detector to the surface while considering the kinematic and visibility constraints of the platform. As the survey requires resilient and accurate localization in diverse terrain, we also propose a factor graph-based online fusion of GNSS, IMU, and LiDAR measurements. We validate the robustness of the solution to individual sensor degeneracy by flying under the canopy of trees and over featureless fields. A simulated ablation study shows that the proposed planner reduces coverage duration and improves trajectory smoothness. Real-world flight experiments showcase autonomous mapping of buried metallic objects in undulated and obstructed terrain.

Hybrid unmanned aerial vehicles (H-UAVs) are highly versatile platforms with the ability to transition between rotary- and fixed-wing flight. However, their (aero)dynamics tend to be highly nonlinear which increases the risk of introducing safety-critical modeling errors in a controller. Designing a safe, yet not too cautious controller, requires a credible model which provides accurate dynamics uncertainty quantification. We present a data-efficient, probabilistic semi-parametric dynamics modeling approach that allows for online, filter-based inference. The proposed model leverages prior knowledge using a nominal parametric model, and combines it with residuals in the form of sparse Gaussian processes to account for possibly unmodeled forces and moments. Uncertain nominal and residual parameters are jointly estimated using Bayesian filtering. The resulting model accuracy and the reliability of its predicted uncertainty are analyzed for both a simulated and a real example, where we learn the 6DoF nonlinear dynamics of a tiltwing H-UAV from a few minutes of flight data. Compared to a residual-free nominal model, the proposed semi-parametric approach provides increased model accuracy in relevant parts of the flight envelope and substantially higher credibility overall.

From construction materials, such as sand or asphalt, to kitchen ingredients, like rice, sugar, or salt; the world is full of granular materials. Despite impressive progress in robotic manipulation of single objects, granular materials remain a challenge due to difficulties in modelling these highly deformable and inhomogeneous materials, which are governed by dynamics that are hard to capture analytically. We argue that despite the high degrees of freedom and the complex underlying dynamics of granular materials, many practical problems that require manipulating them can be solved by leveraging simple models, informative motion priors, and a fast feedback loop. In this work, we show that computational Optimal Transport (OT) can be leveraged to derive informative, robot-agnostic motion priors for transforming a pile of granular materials from a source into a target distribution and generate robot motion plans with a next-best sweep planner that uses a simple material-agnostic sweep model. We plan sweeps directly on a height map representation of the material distribution and hence avoid a costly particle-level treatment of the problem. We validate our approach with a large set of simulation and hardware experiments that demonstrate several complex shaping tasks, including gathering, separating, and writing letters with different types of granular materials.

The field of aerial manipulation has seen rapid advances, transitioning from push-and-slide tasks to interaction with articulated objects. The motion trajectory of these complex actions is usually hand-crafted or a result of online optimization methods like Model Predictive Control (MPC) or Model Predictive Path Integral (MPPI) control. However, these methods rely on heuristics or model simplifications to efficiently run on onboard hardware, limiting their robustness, and making them sensitive to disturbances and differences between the real environment and its model. In this work, we propose a Reinforcement Learning (RL) approach to learn reactive motion behaviors for a manipulation task while producing policies that are robust to disturbances and modeling errors. Specifically, we train a policy to perform a door-opening task with an Omnidirectional Micro Aerial Vehicle (OMAV). The policy is trained in a physics simulator and shown in the real world, where it is able to generalize also to door closing tasks never seen in training. We also compare our method against a state-of-the-art MPPI solution in simulation, showing a considerable increase in robustness and speed.

Autonomously navigating a robot in everyday crowded spaces requires solving complex perception and planning challenges. When using only monocular image sensor data as input, classical two-dimensional planning approaches cannot be used. While images present a significant challenge when it comes to perception and planning, they also allow capturing potentially important details, such as complex geometry, body movement, and other visual cues. In order to successfully solve the navigation task from only images, algorithms must be able to model the scene and its dynamics using only this channel of information. We investigate whether the world model concept, which has shown state-of-the-art results for modeling and learning policies in Atari games as well as promising results in 2D LiDAR-based crowd navigation, can also be applied to the camera-based navigation problem. To this end, we create simulated environments where a robot must navigate past static and moving humans without colliding in order to reach its goal. We find that state-of-the-art methods are able to achieve success in solving the navigation problem, and can generate dream-like predictions of future image-sequences which show consistent geometry and moving persons. We are also able to show that policy performance in our high-fidelity sim2real simulation scenario transfers to the real world by testing the policy on a real robot. We make our simulator, models and experiments available at https://github.com/danieldugas/NavDreams.

Autonomous navigation among people is a com-plex problem that also exhibits considerable variation depending on the type of environment and people involved. Here we consider navigation among crowds that exhibit flow-like behavior like people moving through a train station. We propose a novel pseudo-fluid model of crowd flow for such problems. These have an intuitive physical interpretation and do not require much tuning. We further formalize an observation model to infer flow properties from discrete sensor observations, including support for partial observability, and pair it with a flow-aware planner. We demonstrate the potential of the approach in simulated navigation scenarios. We achieve state of the art results on the CrowdBot navigation benchmark, and also compare favorably against a standard ROS planner on a partially observable environment, demonstrating that the flow-aware planner successfully estimates and plans around counter-flows in the crowd in real time. We conclude that flow-based planning shows great promise for crowded environments that may exhibit such flow-like behavior.

In this work we address the challenging problem of manipulating a flexible link, like a rope, with an aerial robot. Inspired by spraying tasks in construction and maintenance scenarios, we consider the case in which an autonomous end-effector (e.g., a spray nozzle moved by a robot or a human operator) is connected to a fixed point by a rope (e.g., a hose). To avoid collisions between the rope and the environment while the end-effector moves, we propose the use of an aerial robot as a flying companion to properly manipulate the rope away from collisions. The aerial robot is attached to the rope between the end-effector and the fixed point. Assuming no direct control of the end-effector (e.g., when operated by a human), we design a reactive and fast motion planner for the aerial robot. Grounding on the theory of Forced Geometric Fabrics, we design a motion planner that generates trajectories to drive the aerial robot to follow the end-effector, while manipulating the rope to avoid collisions in cluttered environments. To include the complex behavior of the flexible link, we propose a rope model that estimates its real-time state under forces and position-based interactions, as well as collisions with obstacle surfaces. Finally, we evaluate the system behavior and the motion planner performance in simulations, as well as in real-world experiments on an original spray painting application.

Modern optimization-based approaches to control increasingly allow automatic generation of complex behavior from only a model and an objective. Recent years has seen growing interest in fast solvers to also allow real-time operation on robots, but the computational cost of such trajectory optimization remains prohibitive for many applications. In this paper we examine a novel deep neural network approximation and validate it on a safe navigation problem with a real nano-quadcopter. As the risk of costly failures is a major concern with real robots, we propose a risk-aware resampling technique. Contrary to prior work this active learning approach is easy to use with existing solvers for trajectory optimization, as well as deep learning. We demonstrate the efficacy of the approach on a difficult collision avoidance problem with non-cooperative moving obstacles. Our findings indicate that the resulting neural network approximations are least 50 times faster than the trajectory optimizer while still satisfying the safety requirements. We demonstrate the potential of the approach by implementing a synthesized deep neural network policy on the nano-quadcopter microcontroller.

Robots are increasingly expected to move out of the controlled environment of research labs and into populated streets and workplaces. Collision avoidance in such cluttered and dynamic environments is of increasing importance as robots gain more autonomy. However, efficient avoidance is fundamentally difficult since computing safe trajectories may require considering both dynamics and uncertainty. While heuristics are often used in practice, we take a holistic stochastic trajectory optimization perspective that merges both collision avoidance and control. We examine dynamic obstacles moving without prior coordination, like pedestrians or vehicles. We find that common stochastic simplifications lead to poor approximations when obstacle behavior is difficult to predict. We instead compute efficient approximations by drawing upon techniques from machine learning. We propose to combine policy search with model-predictive control. This allows us to use recent fast constrained model-predictive control solvers, while gaining the stochastic properties of policy-based methods. We exploit recent advances in Bayesian optimization to efficiently solve the resulting probabilistically-constrained policy optimization problems. Finally, we present a real-time implementation of an obstacle avoiding controller for a quadcopter. We demonstrate the results in simulation as well as with real flight experiments.

Reinforcement learning for robot control tasks in continuous environments is a challenging problem due to the dimensionality of the state and action spaces, time and resource costs for learning with a real robot as well as constraints imposed for its safe operation. In this paper we propose a model-based reinforcement learning approach for continuous environments with constraints. The approach combines model-based reinforcement learning with recent advances in approximate optimal control. This results in a bounded-rationality agent that makes decisions in real-time by efficiently solving a sequence of constrained optimization problems on learned sparse Gaussian process models. Such a combination has several advantages. No high-dimensional policy needs to be computed or stored while the learning problem often reduces to a set of lower-dimensional models of the dynamics. In addition, hard constraints can easily be included and objectives can also be changed in real-time to allow for multiple or dynamic tasks. The efficacy of the approach is demonstrated on both an extended cart pole domain and a challenging quadcopter navigation task using real data.